The resolving power of the computer- and video-based light microscope is limited by two main factors. (1) It is limited by the optical resolving power of the microscope, which is -0.25 micro meter (2) It is limited by the pixel size. Resolving power of less than 0.2 micro meter usually requires an electron microscope. Three dimensional imaging to resolving powers of less than 0.2 micro meter requires either an intermediate voltage electron microscope or a high voltage electron microscope (HVEM), both of which are not readily available at most institutions. Live imaging is not possible with any electron microscope. We propose a novel blind deconvolution software algorithm that will commercially provide subpixel resolving power of well under 0.2 micro meter for live fluorescence microscope imaging. The software will be affordable to most biology laboratories and will run on the Windows NT/98 platform. It will require no add-on equipment for the PC or the microscope, which is a critical factor for acceptance by customers and original-equipment-manufacturers through whom the software will be sold. Algorithm research and test bed experimentation of the Phase 1 feasibility will involve verification of subdiffraction 3D structure of the centrosome, a fabricated test target and a computer simulation. PROPOSED COMMERCIAL APPLICATION: Applications include 2D and 3D fluorescence microscopy (FM) and transmitted light brightfield microscopy. Live FM will have a large scientific impact, since presently there is no commercial means for resolving power under 0.2 micrometers. Market inquiries by us indicate that on the order of 400 units per year should be sold by us and an aggregate of at least 3 times this number should be sold by inevitable enduring competitors who will develop competitive products after our product is introduced.